View
217
Download
0
Category
Preview:
Citation preview
EMPIRICAL STUDY OF
BLACK-SCHOLES WARRANT PRICING MODEL
ON THE STOCK EXCHANGE OF MALAYSIA
by
HONG BOON KYUN
Research report in partial fulfillment of the requirements
for the degree of Master of Business Administration
MARCH 2004
ii
ACKNOWLEDGEMENTS
Dr. Zamri Ahmad and Dr. Noor Azlinna Azizan, my supervisors. After many hours
of fruitful discussion and consultation, I have learnt a lot from you especially in
developing a finance research. Thank you deeply on your guidance, patience and
precious time spent on this research.
All lecturers. Thank you for leading me to this stage. I have got a real sense of
scientific research in business and finance only here at the School of Management in
USM during my MBA studies.
Mr. Jimmy Lee, my superior colleague. Thank you for offering me your advice,
digging up useful research articles and sharing your knowledge in the areas of
research and finance with me.
Thanks to my parents, brothers and sisters for bringing up me and for being so
supportive throughout and for all their encouragement, inspiration and moral
support.
Last but not least, my patient and loving fiancée, Siew Li, who has been a great
source of strength for me to go through this work. Thank you for being there for
me. I know that I will always have someone that I can lean on.
iii
TABLE OF CONTENTS
Page
TITLE PAGE i
ACKNOWLEDGEMENTS ii
TABLE OF CONTENTS iii
LIST OF TABLES vi
LIST OF FIGURES vii
LIST OF APPENDICES viii
ABSTRAK ix
ABSTRACT x
Chapter 1: INTRODUCTION
1.1 General Introduction 1
1.2 Problem Statement 2
1.3 Research Questions 3
1.4 Research Objectives 4
1.5 Definition of Terms 4
1.6 Significance of the Study 5
1.7 Organization of Remaining Chapters 6
Chapter 2: LITERATURE REVIEW
2.1 Introduction 8
2.2 Practical Review 8
2.2.1 Significance of the Model - Literature Context 9
2.2.2 Significance of the Model - Malaysia Context 9
iv
Page
2.3 Empirical Test on Different Markets 11
2.4 Black-Scholes Model Assumptions 12
2.4.1 Do BS Model Assumptions Hold? 13
2.5 Striking Price Bias 16
2.6 Time To Expiration Bias 17
2.7 Variance Bias 18
2.8 Bias Reported in Different Time Period 18
2.9 Theoretical Framework 20
2.9.1 Variables Affecting the Price of Warrants 20
2.10 Development of Hypotheses 22
2.11 Summary 23
Chapter 3: METHODOLOGY
3.1 Black-Scholes Computational Model 25
3.2 Research Design 26
3.3 Methodology 26
3.3.1 Population and Sample 26
3.3.2 Data Collection Method 29
3.4 Data Analyses 30
3.3.1 Measurement of Volatility 30
3.3.2 Statistical Techniques 32
3.5 Summary 34
v
Page
Chapter 4: STATISTICAL ANALYSES AND RESULTS
4.1 Summary Statistics of the Data Set 35
4.2 Normality Test of Distribution 36
4.3 Test on Regression Model 37
4.4 Overall Price Observations 39
4.5 Testing on Individual Warrants 40
4.6 Pricing Biases 40
4.6.1 Striking Price Bias 41
4.6.2 Time to Expiration Bias 43
4.6.3 Variance Bias 45
4.7 Pre and Post Asian Financial Crisis 47
4.8 Summary 49
Chapter 5: DISCUSSION AND CONCLUSION
5.1 Recapitulation of the Study and Discussion 50
5.2 Limitation of the Study 53
5.3 Suggestions for Future Research 54
5.4 Implications of the Findings 54
5.5 Recommendations and Conclusion 55
REFERENCES 56
APPENDICES 62
vi
LIST OF TABLES
Page
Table 2.1 List of Local Institutions/ Publications that Often Apply
Black-Scholes Model 10
Table 2.2 KLSE Clearing Fees 15
Table 2.3 Changes in Warrant Price against Variables Based on BS Model 20
Table 3.1 Reasons for Excluding Warrants from the Study 28
Table 4.1 Summary of Sample Warrants by Sectors 35
Table 4.2 Summary Results on Tests of Normality 37
Table 4.3 Summary Results of Regression Tests 38
Table 4.4 Wilcoxon Signed Ranks Test Results on Overall Price
Observations 39
Table 4.5 Wilcoxon Signed Ranks Test on Observations Categorised
by Moneyness 42
Table 4.6 Wilcoxon Signed Ranks Test on Observations Categorised
by Time to Maturity 44
Table 4.7 Wilcoxon Signed Ranks Test on Observations Categorised
by Variance of Underlying Shares 46
Table 4.8 Wilcoxon Signed Ranks Test on Observations Categorised
by Pre and Post Crisis 47
vii
LIST OF FIGURES
Page
Figure 2.1 Growing number of warrants (Year 1995-2003) 8
Figure 2.2 KLSE Composite Index showing turbulent market after
1997 Asian financial crisis (Year 1994-2003) 19
Figure 3.1 Historical yield on 3-month treasury bills (Year 1994-2003) 30
viii
LIST OF APPENDICES
Page
Appendix 1 Population List of Listed Warrants 63
Appendix 2 Worksheet Showing Calculation of BS Price from
Sample Warrant – Poly Berhad 67
Appendix 3 SPSS Output on Normality Test 73
Appendix 4 SPSS Output on Wilcoxon Signed Ranks Test on
Overall Price Observations 77
Appendix 5 Summary Results on Individual Warrant Price Observations 78
Appendix 6 SPSS Output on Wilcoxon Signed Ranks Test on
Price Observations Categorised by Moneyness 80
Appendix 7 SPSS Output on Wilcoxon Signed Ranks Test on
Price Observations Categorised by Time to Maturity 82
Appendix 8 SPSS Output on Wilcoxon Signed Ranks Test on
Price Observations Categorised by Variance 83
Appendix 9 SPSS Output on Wilcoxon Signed Ranks Test on
Price Observations Categorised by Before and After Crisis 85
Appendix 10 SPSS Output on Regression Test 86
ix
ABSTRAK
Kajian ini adalah untuk mengenalpasti setakat manakah ketepatan model penentuan
nilai harga waran/ opsyen yang terkenal – model Black-Scholes – di bursa saham
Malaysia. Dapatan kebanyakan kajian terdahulu (Rubinstein, 1981; Geske, Roll, &
Shastri, 1983; Scott, 1987) terhadap model Black-Scholes adalah positif dan
mendapati model ini berupaya meramalkan harga pasaran waran dengan tepat.
Walau bagaimanapun, didapati ada juga kajian yang memaparkan kegagalan model
ini (Macbeth & Merville, 1980; Lauterbach & Schultz, 1990). Model ini didapati
terlebih menilai waran yang ‘in-the-money’ dan sebaliknya didapati terkurang
menilai waran yang ‘out-of-the-money’. Tujuan kajian ini adalah untuk menguji
ketepatan and kegunaan model Black-Scholes ini di pasaran waran yang lebih kecil
dan kurang cair – Bursa Saham Kuala Lumpur (BSKL). Kajian ini juga
menimbangkan pelbagai kecondongan penentuan harga ke atas harga guna hak
waran, tempoh matang waran, kemeruwapan harga, sebelum dan selepas krisis
kewangan Asia. Kajian ini dijalankan dengan memerhatikan harga harian dari 74
waran sampel dalam tahun 1994-2003. Keputusan kajian ini medapati bahawa harga
model adalah jelas sekali lebih kurang daripada harga pasaran dan seterusnya
memaparkan kedua-dua harga model dan harga pasaran adalah berbeza secara
bersistematik ke atas pelbagai kecondongan penentuan harga tersebut. Sebagai
kesimpulan, pengguna model Black-Scholes perlu berhati-hati dan memerhatikan
corak perbezaan harga secara bersistematik ini apabila memilih sesuatu waran untuk
tujuan pelaburan di bursa saham Malaysia.
x
ABSTRACT
This paper addresses the question of how well the best-known warrant/ option
pricing model – the Black-Scholes model – work in the stock exchange of Malaysia.
Results of most studies (Rubinstein, 1981; Geske, Roll, & Shastri, 1983; Scott,
1987) have been positive in that the Black-Scholes model generates warrant values
fairly close to the actual prices at which warrants trade especially for shorter term
maturity warrants. Nevertheless, some regular empirical failures of the model have
been noted (Macbeth & Merville, 1980; Lauterbach & Schultz, 1990). The Black-
Scholes model tends to overvalue ‘in-the-money’ warrants and undervalue ‘out-of-
the-money’ warrants. The objective of this paper is to test the warrant market
behavior in relation to the application of Black-Scholes model to a relatively small
and less liquid market – Kuala Lumpur Stock Exchange (KLSE). This study
considered various pricing biases related to warrant strike price, time to maturity,
volatility, and pre- and post-Asian financial crisis period. This paper has tested the
model using daily prices of 74 sample warrants in the year 1994-2003. The results
revealed that overall model prices were significantly below market prices and
further indicated both the model and market prices deviate in certain systematic
patterns as for the above various pricing biases. It was concluded that users of
Black-Scholes model should carefully observe the systematic pattern of deviation
when choosing an investment of warrants in the Malaysian stock exchange.
1
Chapter 1
INTRODUCTION
1.1 General Introduction
A warrant is an investment where the warrant issuer offers an option to the warrant
holder, the right but not the obligation, to buy new ordinary shares from the issuer at
a predetermined price at any point within this given time period. The definition of
warrant is very close to that of a call option especially in term of valuation of its fair
price where it is common to value warrants using option pricing models, such as
Black-Scholes model. There are enormous empirical studies of Black-Scholes (BS)
warrant pricing model that include Galai and Schneller (1978), Macbeth and
Merville (1980), Rubinstein (1985), Geske, Roll, and Shastri (1983), Leonard and
Solt (1990). The results in these studies regarding the applicability of the
Black-Scholes model to warrant pricing are mixed. The Black-Scholes model is
commonly found to perform as good as and is often more accurate than other
warrant pricing model (Leonard & Solt, 1990). Nevertheless, some empirical
failures of the model have been noted (Macbeth & Merville, 1980). General
empirical findings revealed that the model tends to overvalue in-the-money warrants
and undervalue out-of-money warrants. Somehow, the model probably brings
forward a great impact to Malaysian equity warrant investors since it is widely used
as equity warrant evaluator by local investment analysts. In order to answer the
questions as to its applicability and reliability, this paper focuses on examining the
market pricing behavior of listed warrants against fair value produced by the model.
The empirical result of this study shall serve as evidence as to what extent the
Black-Scholes model has performed on the stock exchange of Malaysia.
2
1.2 Problem Statement
Warrant is relatively more volatile as compared to its underlying share. To a certain
extent, warrants offer better upside potential and tend to rise faster, in percentage
terms, than their underlying share – the higher the gearing, the faster the price
increases. However, in a bear market, everything works in reverse. Furthermore,
the market for warrants in Malaysia is still small, hence warrants may be volatile
and become targets for artificial rallies (Ariff, Chan, & Johnson, 1995).
Warrants is more complex as pertaining to its exercise price, conversion ratio,
maturity date, exercise procedures, etc. The complex nature of warrants raise an
important problem statement – what is the fair pricing of warrants? As for many
uneducated investors, these are their principal disadvantages. Without a fairly
workable warrant pricing model, investors may view this as an obstacle and stay
away from warrant market.
Investors who have little grasp and understanding of the nature of warrants can
rarely make good profit. Malaysian traded warrants are relatively new (since 1994)
and many retail investors are not really well equipped with knowledge of warrant
investment. As a result, retail investors can hardly make good profit or achieve their
respective level of expected return.
Black-Scholes model is widely used by local researchers in Malaysia. Investors
may over-relied solely on this model without referring to other common ratio and
financial indicators. Wrong decision can be made while choosing a warrant
investment.
3
Retail and institutional investors may also desire to find out whether the
Black-Scholes model can provide a fairly accurate estimation for warrants with
different length of maturity and degree of riskiness (volatility) of respective
underlying shares, and for in-the-money versus out-of-money warrants. Moreover,
investors may be interested to grasp the recent performance of the model after the
Asian financial crisis (high-interest period) as compared to period before the crisis
(low-interest period). This is to assist investors in making detail evaluation prior to
choosing an investment.
1.3 Research Questions
From the above, general research question is raised as to whether the prices paid by
investors in the market were fair prices should they be valued by the Black-Scholes
model. The specific research questions can be summarized as follows:
1) Is the Black-Scholes model appropriate for pricing of the warrants traded on
the KLSE?
2) Is the Black-Scholes model reasonably accurate for in-the-money versus
out-of-the-money warrants pricing?
3) Is the Black-Scholes model more appropriate to value long-maturity warrants
against near-maturity warrants?
4) Can the Black-Scholes model provide a fairer price for warrants with high
underlying volatility, or low underlying volatility?
5) Does the Black-Scholes model work better during post-Asian financial crisis
period compared to pre-Asian financial crisis period, particularly in
Malaysia?
4
1.4 Research Objectives
The objective of this study is to address the above research questions. The questions
are addressed by performing empirical testing on the performance of Black-Scholes
model that commonly used for warrants traded in developed markets as applied to
those traded on the KLSE. This focus of this empirical study is to compare the
differences between actual market prices and BS model prices. Should the model
performs poorly, this study will analyse all probable pricing biases rather than look
into other alternatives. Probable pricing biases include warrant strike price, time to
maturity, volatility of respective underlying share, and risk-free interest rate. The
Black-Scholes model may over or under perform based on respective probable
pricing biases. As a result, this study will find out whether the model and observed
market prices deviate in certain systematic patterns as for the pricing biases.
1.5 Definition of Terms
For the purposes of this research, the following terms need clarification.
An Option. The right but not an obligation, to buy or sell a particular good at an
exercise price, at or before a specified date.
Call Option. The right but not an obligation to buy a particular good at an exercise
price.
Put Option. The right but not an obligation to sell a particular good at an exercise
price.
Warrants. A long-term call options written on a specified number of ordinary shares
by issuer company and traded on an organized exchange.
Call Warrants. A relatively new type of financial instrument introduced to the
KLSE. A short-term call options (not more than 2 years) that can be issued by a
5
third party entity specified in the Security Commission guidelines based on existing
shares.
Exercise Price or Strike Price. The fixed price at which the good may be bought or
sold.
In the money. Gain from immediate exercise. Positive intrinsic value.
Out of the money. An option that is not worth exercising. Zero intrinsic value.
At the money. Where the share price is equal to the exercise price.
American Option. An option that can be exercised on any day up until its expiry
date.
Pseudo-American Option. An option that can be exercised on a future specified date
up to its expiry date.
European Option. An option that can be exercised on the last day of the option.
Premium. The cost of an option.
Black-Scholes Model. An option pricing model developed by Fischer Black and
Myron Scholes to produce an expected value of a stock option, taking into account
various factors like underlying share price and volatility, warrant exercise price and
time to maturity, and risk-free rate.
1.6 Significance of the Study
The applicability of the Black-Scholes model to warrant pricing is an empirical
issue. Although the model postulated by Black and Scholes (1973) has been widely
tested in developed countries of the West and Japan, studies of the Black-Scholes
model in emerging markets have been relatively scarce. Price inefficiency exists
probably because of the market is under-researched, and not only because of the
market is small, less developed, and less liquid (Ariff et al., 1995).
6
To date, these research questions pertaining to the performance of the model on
KLSE remain unanswered. It is hoped that this study would contribute to the
empirical literature on the valuation of warrants traded at the emerging markets, as
well as stimulate further investigation in the valuation of warrants and other
derivatives traded on KLSE.
According to Restaino and Andrew (2001), directors should understand the
Black-Scholes model in order to determine the economic value of stock option in
assessing and calibrating their own organization’s stock compensation programs. In
view of this, it is also hoped that the results of this study will be of value to directors
and investors to understand the acceptability of Black-Scholes model in determining
the economic value of stock option in Malaysia. As a result, warrant investors will
not follow the model value blindly, as commonly advised by local financial analysts.
In a nutshell, this study is expected to educate investors to have a better
understanding of warrants as hedging, arbitraging, investing or speculative vehicle,
in case of mispricing.
1.7 Organization of Remaining Chapters
The chapters in this study are organized in the following sequence. Chapter 1
provides the introduction, states the nature of problem and research questions that
lead to research objectives. Chapter 2 reviews the relevant previous empirical
research and highlights research gap in some warrant pricing biases, and describes
the theoretical framework of this study and hypotheses to be tested. Chapter 3
explains the research methodology and statistical procedures in the study. Chapter 4
7
reports and analyzes the empirical findings for this study. Last but not least, Chapter
5 sums up the conclusion, states the limitation of the study as well as suggestions for
future research.
8
Chapter 2
LITERATURE REVIEW
2.1 Introduction
A review of related literature is conducted to compare the knowledge of financial
economists, compensation professionals, option traders, academics and other experts
in the area of warrant pricing model evaluation, and assessment of BS model as to
pricing of warrants. The review will serve as a method of comparing and contrasting
the views of experts in the field, and to provide a conceptual framework for the
study.
2.2 Practical Review
Warrants were first listed and traded on the Kuala Lumpur Stock Exchange (KLSE)
in 1994. As at 31st June 2003, there are 183 warrants traded on the KLSE.
Referring to Figure 2.1 below, warrants have grown rapidly in terms of market
capitalization and number of warrants listed.
183178163
153
120117
8372
116
0
50
100
150
200
1995 19961997 1998 19992000 20012002 2003
Nu
mb
er
of
Lis
ted
Wa
rra
nts
0
2
4
6
8
10
12
14
16
18
(RM
Bil)
Number of
WarrantsNominal Value
Figure 2.1. Growing number of warrants (Year 1995-2003).
(Source: KLSE Daily Diary)
9
The increasing popularity raises an important issue – what is the fair pricing for
warrants? The Malaysian market is now in demand of a workable pricing model to
guide investor in warrants investment.
2.2.1 Significance of the Model – Literature Context
Black and Scholes (1973) claimed that in many cases their famous model could be
used as an approximation to give an estimate of the warrant value. The Chicago
Board Options Exchange (CBOE), the first public options exchange, began trading
in April 1973, and by 1975, traders on the CBOE were using the model to price and
hedge their option positions. Since then, “thousands of traders and investors use the
formula everyday”, noted by the Nobel committee (Marsh & Kobayashi, 2000). It
was widely used in those personal computer days that Texas Instruments sold a
handheld calculator specially programmed to produce Black-Scholes options prices
and hedge ratios.
Merton (1998) remarked that the influence of the Black-Scholes option theory on
finance practice has not been limited to financial options traded in markets or even
to derivatives generally. It is also used to price and evaluate risk in a wide array of
applications, both financially and non financially. He further remarked that the
publication of the option model in 1973 surely helped the development and growth
of the listed options and over-the-counter (OTC) derivatives markets.
2.2.2 Significance of the Model – Malaysian Context
In the Malaysian context, Malaysia Derivatives Exchange Bhd (MDEX) and
KLSE-RIIAM Information System (KLSE-RIS) has inserted a Black-Scholes
10
Option Pricing Model Calculator in their respective website for investors
convenience in checking out BS price. KLOFFE (before MDEX) has become the
first exchange to provide Black-Scholes Model calculator over internet in Southeast
Asia, beginning 16th
April 2001. Table 2.1 shows a list of local institutions that
often apply the Black-Scholes Model to value Malaysian warrants and advise
investors. The model brings forward a great impact to Malaysian equity warrant
investors since it is widely used as equity warrant evaluator by local research
institutions and securities firms.
Table 2.1
List of Local Institutions/ Publications that Apply Black-Scholes Model
TA Securities RHB Research Institute
Kenanga Research Affin Securities
AmSecurities iCapital.biz
GK Goh Research Malaysian Business
AMS Research Dynaquest's Monthly Digest
(Source: The Edge and other local research articles)
One of the conditions for Securities Commission’s (SC) approval for proposed
replacement warrants is that the replacement warrants to be issued must be at a fair
value based on the Black-Scholes valuation model. This has been done to Naluri
Bhd and Minho (M) Bhd replacement warrants. As a whole, the application of the
11
Black-Scholes model in Malaysia is significance and hence the research questions
pertaining to the applicability of the model should not be overlooked.
2.3 Empirical Test on Different Markets
There have been an enormous number of empirical tests on the Black-Scholes
model. The size of the warrant or option market, and especially the difference in
size between the U.S., Japan as well as emerging markets in Asia, and Malaysia in
particular is interesting from the viewpoint of researchers. Empirical researches on
call options traded on developed market in the U.S. include Macbeth and Merville
(1980), Rubinstein (1985), Lauterbach and Schultz (1990), Leonard and Solt (1990),
and Hauser and Lauterbach (1997). Macbeth and Merville (1980) and Rubinstein
(1985) reported actual call option prices on the U.S. market tend to trade above their
Black-Scholes values when the options are out-of-the money and below their
Black-Scholes values when the options are in-the-money.
Kuwahara and Marsh (1992) investigated how well the prices of Euro-equity
warrants issued by Japanese companies can be explained by the Black-Scholes
model. They found similar pattern of discrepancies between Japanese equity
warrant prices and Black-Scholes values to that was reported in previous studies
(Macbeth & Merville, 1980; Rubinstein, 1985) of the performance of the
Black-Scholes model in pricing U.S. stock options, at least over selected periods of
time.
On the other hand, Huang and Chen (2002) have examined the performance of the
Black-Scholes model applied to the valuation of covered warrants traded on the
12
Taiwan Stock Exchange (TSE) while Duan and Yan (1999) have investigated the
market pricing of derivative warrants written on the HSBC common stock traded on
the Hong Kong Stock Exchange. Interestingly, both Huang and Chen (2002) and
Duan and Yan (1999) have empirically shown that the model underpriced the
warrants in respective market. The conclusion reached in both of their papers seems
inconsistent with the general empirical findings about the Black-Scholes model.
Empirical research on warrants traded on emerging markets in Southeast Asia
include the work of Shastri and Sirodom (1995), who concluded that a constant
elasticity of variance model outperformed Black-Scholes model in pricing Thailand
warrants. Santoso (2000) has investigated empirically the relative performance of
pricing models for warrants traded at the Jakarta Stock Exchange (JKX) and found
out that the Black-Scholes and Dividend-adjusted Black-Scholes are doing equally
well. For the most part, the results of the studies have been positive in that the
Black-Scholes model generates option values fairly close to the actual prices at
which options are traded. Nevertheless, different test results could have been
produced in different markets.
2.4 Black-Scholes Model Assumptions
The version of Black-Scholes model is predicated based on a few simplifying
assumptions (Black & Scholes, 1973).
I. The stock price follows a random walk in continuous time with a
variance rate proportional to the square of the stock price. Thus the
distribution of possible stock price at the end of any finite interval is
log-normal. The variance rate of the return on the stock is constant.
13
II. The short-term interest rate is known and is constant through time.
III. The option is “European”, that is, it can only be exercised at maturity.
IV. The stock pays no dividend or other distributions.
V. There are no transaction costs in buying or selling the stock or the option.
VI. It is possible to borrow any fraction of the price of a security to buy it or
to hold it, at the short-term interest rate.
VII. There are no penalties to short selling. A seller who does not own a
security will simply accept the price of the security from a buyer, and
will agree to settle with the buyer on some future date by paying him/her
an amount equal to the price of the security on that date.
2.4.1 Do Black-Scholes Model Assumptions Hold?
Central to the Black-Scholes model is the assumption that the underlying stock price
moves randomly, following a geometric Brownian motion process (Ariff et al.,
1995). Kim, Oh, and Brooks (1994) argued that this would allow the price of an
option to be computed from an estimation of the price variances of the underlying
asset that can be readily obtained from the sample. However, Merton (1976) and
Fortune and Peter (1996) found that the stock price distribution did not conform
strictly to the normality assumption.
Fortune and Peter (1996) commented that the Assumption No II was adopted for
convenience and not strictly true. However, this assumption was common to finance
theory, for example it was one of the assumptions of the Capital Asset Pricing
Model. Malaysian short-term risk-free interest, which is obtainable using 3-months
Malaysian Treasury Bills yield, was reasonably flat since May 1999 until June 2003
14
(after Asian financial crisis) at the range of 2.65%-2.96%. However, the rate may
fluctuate exceeding the range under certain market conditions.
Even though Malaysian equity warrants are pseudo-American where it can be
exercised at any point within the time period or future period (as set out in Deed
Poll) prior to its maturity, the European option assumption serves as the foundation
in most of all options formula used today. Fortune and Peter (1996) explained the
ability to focus on an European-style option for which the Black-Scholes model was
designed would allowed us to ignore the potential influence of probable early
exercise.
Some of the underlying shares do pay dividend or distribute bonus. However, prices
of warrants and underlying shares, that are applied in Black-Scholes computational
model, will be adjusted on ex-dividend dates by the amount of dividends paid, bonus
issues or share splits in order to meet Assumption No IV and assist in price
comparison.
Furthermore, the transaction costs in buying or selling the stocks or the warrants in
Malaysia is very minimal. Referring to Table 2.2, KLSE clearing fees for on-market
transaction (buying or selling) is only at 0.04% of transaction value. Also, there is
no tax charged on any capital gains made upon the disposal of derivatives or
securities in Malaysia. These have made the empirical study of Malaysian equity
warrants pricing by using Black-Scholes model feasible.
15
Table 2.2
KLSE Clearing Fees
On-Market 0.04% of transaction value (payable by both buyer and seller) with a maximum of RM200.00 per contract. No minimum.
Direct Business 0.04% of transaction value (payable by both buyer and seller), with a maximum of RM200.00 and a minimum of RM10.00.
Institutional Settlement Service
RM25.00 - Non-Trading Clearing Member who issues an ISS Confirmation/Affirmation.
(Source: KLSE Website, http://www.klse.com.my/website/trading/settlement.htm.)
The Black-Scholes model is derived under the usual perfect-market assumptions,
such as the securities are perfectly divisible and continuously tradable without
transactions costs, taxes and restriction on short sales. Guo (1999) commented that
should none of these assumptions were accepted, most of the finance theories would
have to be rejected. Hence, the only possibility to challenge the Black-Scholes
model is to argue the validity of its assumption on the underlying process.
The results in these studies regarding the applicability of the Black-Scholes model to
warrant valuation are mixed. Leonard and Solt (1990), and Kremer and Roenfeldt
(1993) concluded that the Black-Scholes model was commonly found to perform as
well as more complicated alternative models for warrant pricing. On the other hand,
Lauterbach and Schultz (1990), followed by Shastri and Sirodom (1995), and
Hauser and Lauterbach (1997) presented empirical evidence that suggested the
Black-Scholes model was outperformed by a model that assumed a constant
elasticity of variance (CEV) diffusion process for stock price. However, Guo (1999)
has commented the existing literature has not been conclusive on whether the
Black-Scholes model, including all the alternative solutions, is wrong. Any theory
16
that gives best prediction consistently is the best theory, regardless of the
assumptions required to generate them (Fortune & Peter, 1996).
2.5 Striking Price Bias
Black (1975) defined forward price equal to the strike price for at-the-money
warrants and assumed that the current market price is the present value of the
forward price. Nevertheless prior researches (Galai & Schneller, 1978; Hull &
White, 1987) have suggested that the Black-Scholes model is a poor predictor of
in-the-money and out-of-money options while it is a highly effective predictor of
at-the-money options. Lauterbach and Schultz (1990) documented that the
Black-Scholes model, as adopted for warrants, consistently misprices warrants –
particularly out-of-money warrants. Black (1975) reported that the model
systematically underpriced deep out-of-money options while it overpriced deep
in-the-money options during the 1973-1975 period. Macbeth and Merville (1980)
confirmed the existence of a striking price bias but contrary to Black’s (1975)
findings. Macbeth and Merville (1980) and Shastri and Sirodom (1995) reported the
Black-Scholes model underpriced in-the-money options, overpriced out-of-money
options, and gave an approximate and proper price for options at-the-money when
the stochastic process that generating the stock price was a constant elasticity of
variance process. Kuwahara and Marsh (1992) noticed the same pattern of
discrepancies for Nikkei 225 options with those reported by Black earlier.
Shastri and Sirodom (1995) outlined the possible reasons for the model to
underprice in-the-money warrants and overprice out-of-the-money warrants in that it
did not properly incorporate the fact that market participants were not allowed to
17
short sell securities in Thailand. Otherwise, these discrepancies may be an indicator
of an underlying stochastic process (volatility of the rate of return of the firm value)
that differs from lognormal (Lekkas, 2002).
2.6 Time To Expiration Bias
Since warrants have much longer maturity than options, greater mispricing may be
resulted from an incorrect assumption that stock price volatility is constant at
historical level, and interest rate is known and constant over the life of the warrants.
A distant expiration challenges the reasonableness of the Black-Scholes model.
Perhaps Black and Scholes did not detect this longer maturity “bias” because in their
study they used options at time of issue, when there were 6 months until expiration
(Geske et al., 1983).
While Leonard and Solt (1990) analysed valuation errors for warrants, they
categorized those warrant with time to expiration less than 2 years, between 2-4
years, and more than 4 years. They reported the Black-Scholes model tends to
systematically undervalue out-of-money warrants with less than two years maturity.
However, Ariff et al. (1995) investigated one of the very liquid call contract – Sime
Darby of Stock Exchange of Singapore (SES) and found that the tests on difference
between model and market prices were not highly significant throughout the period
irrespective of the maturity or year of trading.
18
2.7 Variance Bias
Black and Scholes (1972) found that the market appeared to underestimate the effect
of differences in variance rate on the value of an option. They reported the
difference between the price paid by option buyers and the value given by the
formula is greater for options on low-risk (variance) stocks than for options on
high-risk (variance) stocks. They further found that using past data to estimate the
variance has caused the model to overprice options on high variance stocks and
underprice options on low variance stocks. Geske et al. (1983) demonstrated that
these systematic “biases” should be expected when using the Black-Scholes model
to value relatively shorter-term over-the-counter (OTC) options. Geske and Roll
(1984) believed that non-stationary of stock volatilities could be a possible source
for the variance bias.
2.8 Bias Reported in Different Time Period
Rubinstein (1994) found that strike price biases using the Black-Scholes model were
statistically significant but the direction of the observed biases changed from one
period to another. Black (1975) reported that the model underpriced deep
in-the-money options using Chicago Board Options Exchange (CBOE) during the
1973-1975 period. Using CBOE data, MacBeth and Merville (1980) confirmed the
existence of a striking price bias for the 1975-1976 period, but found that it was the
reverse of the bias reported by Black (1975). Rubinstein (1985) also found that the
striking price bias had reversed in the 1976-1977 period. Nevertheless, Rubinstein
(1985) and MacBeth and Merville (1980) reported that the original striking bias
observed by Black (1975) had reestablished itself in late 1977 and 1978. Using
same market data, Whaley and Gray (1997) also found striking price,
19
time-to-maturity and variance biases for the period 1975-1978 but did not report on
the reversal of the striking price bias.
The Asian financial crisis began in 2nd
July 1997 when Thailand announced a
managed float of the baht after using up $33 billion in defending the currency.
Since then, Malaysian stock markets crumbled into a Composite Index level near
260 and appeared turbulence after the crisis (refer to Figure 2.2). Also, the yield on
3-month Malaysian treasury bills had been reduced from 6.76% (1997) to a low of
2.71% (1999). Duan and Yan (1999) divided their sample into two sub-periods with
the data before and after 14th
August 1997, to study the attributes of Black-Scholes
model caused by the Asian financial crisis. They have found that the mean
percentage errors (MPE) and mean absolute percentage errors (MAPE) after the
financial crisis were particularly high. Furthermore, Ariff et al. (1995) considered
the first 3 years since call options were introduced in the Stock Exchange of
Singapore (SES) as learning period and found out that after the learning period
(post-learning period) the market prices appeared to have converged to fair prices.
Figure 2.2. KLSE Composite Index showing turbulent market after 1997
Asian financial crisis (Year 1994-2003).
02/07/1997 Asian financial crisis
20
2.9 Theoretical Framework
Financial economists searched for years for a workable option-pricing model before
the now famous Black-Scholes formula for valuing options published in 1973 by
two financial academics, the late Fischer Black and Myron Scholes and
co-developed by Robert Merton. Scholes and Merton shared the 1997 Nobel Prize
in economics for their accomplishment. It is now widely used by options market
participants.
2.9.1 Variables Affecting the Price of Warrants
The Black-Scholes model views the price of a warrants as a function of the time to
expiration, the price of the underlying security, the exercise price, the risk-free rate
of return, and the instantaneous variance of the security’s return. The five main
variables that influence the value of warrants are outlined in Table 2.3.
Table 2.3
Changes in Warrant Price against Variables Based on BS Model
Increase in Changes in Warrant Price
Underlying Share Price Increase
Exercise Price Decrease
Volatility – σ Increase
Time To Expiry Increase
Risk-free Rate Increase
21
The impact of the above variables on the value of the warrant can be seen by
examining each variables of the Black-Scholes model. An increase in stock price
today always increases the value of a warrant. Given that the exercise price of a
warrant is fixed, any increase of the underlying share price should accommodate
with an increase of the warrant price as a warrant’s value in layman sense is equal to
the share price minus the exercise price. On the other hand, if the share price is
down under the exercise price and it is almost certain to expire without being
exercised, its value will be near zero and effectively worthless.
Exercise price is fixed and does not change throughout the warrant lifespan unless it
has a provision of step-up exercise price as set out in Deed Poll. A high (low)
exercise price will reduce (increase) the fair value of a warrant. Volatility is a
standard measurement of risk on the underlying shares and is the price movement of
the underlying stock. Higher volatility indicates higher movement of the underlying
share price (i.e. higher risk) and this potential upward price movement would induce
higher prices for warrants.
The shorter the time to expiry, or called time decay, the lower the value of warrants.
When the expiry date is soon approaching, the value of the warrants that
approximately equal to the stock price minus the exercise price shall be near zero if
out-of-money. This is because the warrant has less time to perform to breakeven or
move into-the-money.
22
An increase in the interest rate or risk-free rate increases the value of a warrant.
This is because an increase in interest rate will eventually increase the cost of fund
carry and the warrants price must increase to cover the increased cost of carry.
2.10 Development of Hypotheses
There have been an enormous number of empirical tests of the Black-Scholes
model. For the most part, the results of the studies have been positive in that the
Black-Scholes model generates warrants values fairly close to the actual prices at
which warrants trade. Nevertheless, some regular empirical failures of the model
have been noted. Prior research (Galai & Schneller, 1978; Hull & White, 1987;
Lauterbach & Schultz, 1990) have suggested that the Black-Scholes model is a poor
predictor of in-the-money and out-of-money options while it is a highly effective
predictor of at-the-money options. While Geske et al. (1983) demonstrated there are
systematic variance biases when using the Black-Scholes model to value OTC
options. Thus, the study is to test the applicability of the model to warrant pricing in
Malaysia market. It can be examined by comparing the means of both market value
and Black-Scholes value generally and re-grouped it according to different degree of
moneyness, length of maturity, high-low variances, and pre- and post-Asian
financial crisis to test its effectiveness respectively. As a result, the following
alternate hypotheses are developed.
H1,A : The observed market value on all warrants ≠ The BS value on all warrants
H2(i),A : The market value for in-the-money warrants ≠ The BS value for
in-the-money warrants
23
H2(ii),A : The market value for out-of-money warrants ≠ The BS value for
out-of-money warrants
H3(i),A : The market value for near-maturity warrants ≠ The BS value for
near-maturity warrants
H3(ii),A : The market value for far-maturity warrants ≠ The BS value for far-maturity
warrants
H4(i),A : The market value for warrants with high-variance underlying stocks ≠
The BS value for high-variance underlying stocks
H4(ii),A : The market value for warrants with low-variance underlying stocks ≠
The BS value for low-variance underlying stocks
H5(i),A : The market value on all daily price observations before Asian financial
crisis ≠ The BS value on all daily price observations before Asian financial
crisis
H5(ii),A : The market value on all daily price observations after Asian financial crisis
≠ The BS value on all daily price observations after Asian financial crisis
2.11 Summary
Results in these studies regarding the applicability of the Black-Scholes model to
warrant pricing is mixed. Experts in the field of financial economics have
acknowledged the importance of Black-Scholes model. The model is well known to
be able to price at-the-money warrants quite fairly or about equal to the market
prices especially for shorter-term maturity warrants. Leonard and Solt (1990),
Kremer and Roenfeldt (1993) concluded that the model performs as good as
alternative more complicated models for warrant pricing. On the other hand, there
are some regular empirical failures and limitations of the model have been noted.
24
MacBeth and Merville (1980), Rubinstein (1985), Whaley and Gray (1997), and
even Black (1975) himself have acknowledged the existence of striking price,
time-to-maturity and variance biases in this model. From the above, the general
null hypothesis is: There is no significant difference between the market value of
warrants and the Black-Scholes value of warrants.
Recommended